AI memory boosts advanced process demand; HBM high-layer stacking drives precision measurement business opportunities * 2026.10.28 * 17:36 * Commercial Times, Chen Chih-hsiung Demand for artificial intelligence (AI) computing continues to drive capacity expansion for high-bandwidth memory (HBM). As memory chips move toward stacking technologies with more than 12 layers, semiconductor manufacturing is also raising requirements for grinding processes, material formulations, and quality stability. Precision measurement equipment provider Mettler Toledo (METTLER TOLEDO) noted that as advanced packaging processes become increasingly complex, in addition to wafer grinding equipment itself, the formulation accuracy and supply stability of front-end chemical mechanical polishing slurry are gradually becoming important links affecting process consistency, driving demand for high-precision weighing, tank monitoring, and digital process management. HBM improves memory bandwidth and data transmission efficiency by stacking multiple layers of DRAM dies, and has become an important component of AI accelerators. As the number of stacked layers increases, precision requirements for die thickness, surface flatness, and interconnect structures become stricter, bringing new challenges to chemical mechanical polishing (CMP) processes. In particular, in steps such as through-silicon via (TSV) reveal, wafer backside planarization, and subsequent die bonding, thickness non-uniformity, over-polishing, or surface defects during polishing may affect subsequent stacking quality and increase process variation and yield risk. Li Pei-ching, manager of Mettler Toledo's industrial weighing division, said that CMP process stability depends not only on polishing equipment and process parameters; the consistency of raw material formulation and supply of slurry is equally important. Because slurry contains abrasive particles and various chemical additives, deviations in raw material input, dilution ratio, and additive concentration may further affect material removal rate, wafer surface uniformity, and defect control. For high-layer-count HBM processes, maintaining quality consistency of slurry across different batches has become an important issue in material supply and process management. Li pointed out that traditional process management mostly focuses on parameter control at the equipment end, but as semiconductor process precision requirements rise, metering errors in the material formulation stage also need to be included in management. Through high-precision industrial weighing equipment and digital monitoring systems, raw material input and tank weight changes can be tracked in real time, and batch formulas and process records can be established, helping manufacturers identify anomalies early and reducing process risks arising from inconsistent material proportions. In practical applications, Mettler Toledo recently participated in an expansion project of a global memory manufacturer, helping it improve its existing tank weighing and process data management systems. During capacity expansion, the customer faced issues such as tight equipment resources, insufficient anomaly warnings, and cross-system data integration, so it introduced digital load cells, high-precision feeding equipment, and process data management technologies to strengthen monitoring capabilities in slurry formulation and supply. Mettler Toledo said such systems can continuously monitor tank weight and raw material input changes and integrate the relevant data into manufacturing management systems for engineers to track batch records and anomalies. For semiconductor processes that use large amounts of high-value chemical materials, accurately tracking feeding and inventory changes helps reduce material loss risk and improve the traceability of process information. In addition to measurement accuracy, equipment installation and calibration are also factors affecting data reliability. Mettler Toledo pointed out that if industrial weighing systems are affected by mechanical stress, installation deviation, or long-term operating environments, measurement results may drift. Therefore, from load cell configuration and equipment calibration to data transmission, a consistent management mechanism must be established to meet semiconductor production lines' requirements for long-term operation and process stability. As HBM and advanced packaging technologies continue to develop, demand for semiconductor equipment is gradually extending from main process tools to material supply, automated monitoring, and process data management. Mettler Toledo said it will continue to invest in high-precision industrial weighing and digital measurement technologies in the future, helping semiconductor manufacturers improve material formulation, equipment monitoring, and data traceability capabilities to meet the increasingly higher requirements of advanced processes for precision manufacturing and quality management.

Eagle launches SimulTask 6 at Pack Expo. SimulTask 6 provides real-time production statistics and recorded reject reasons to help processors identify changes and investigate rejected products. October 26, 2026 Eagle by Mettler Toledo launched SimulTask 6 at Pack Expo International 2026, held Oct. 18-21 at Chicago's McCormick Place. Built on Eagle's reliable inspection intelligence, the new platform supports AI-enabled, task-based solutions tailored to customer-specific X-ray inspection applications, alongside advanced image analysis, real-time production visibility and automated reporting. At Pack Expo, Eagle demonstrated an AI bone classification model for the Maximizer RMI that identifies bone type using key characteristics such as shape, location and size. By providing real-time information to processing equipment, this task-based solution enables on-the-fly equipment adjustments that can improve production efficiency and product yield. Looking for a reprint of this article? From high-res PDFs to custom plaques, order your copy today! JOIN TODAY to unlock your recommendations. Already have an account? Sign In

Mettler-Toledo International (MTD) shares have climbed 13.01% in a month and 26.93% in three months, ahead of an earnings report expected on November 5, 2026. The company has averaged a 4.36% earnings beat in recent quarters. At a $1,684.64 close, the stock trades at a price-to-earnings ratio of 37.3x, well above an estimated fair ratio of 23.1x. It is cheaper than the Life Sciences sector average of 42.2x and the peer average of 41.5x. A discounted cash flow model values the shares at $1,411.07, which suggests they are overvalued. Earnings are forecast to grow 7.01% a year, slower than the broader US market.